ジオシンセティックス論文集

第34巻(2019.12)

Effect of Cement Solidification Agent and Slurry

Density on Strength and Deformation Property for
Liquefied Stabilized Soil Reinforced by Fiber
Material
Yujie CUI1, Yukihiro KOHATA2

The Liquefied Stabilized Soil (LSS) is easy to occur brittle failure with the increase of cement content and it was found
that the addition of fiber content might improve the brittle property of the LSS. In this study, a series of triaxial
compression tests were conducted with LSS specimens prepared in different slurry densities of 1.216, 1.280 g/cm3,
cement content of 80, 100 kg/m3 and the fiber content of 0, 10 kg/m3. Based on the results of the triaxial test, the
influence of changing in slurry density, as well as added amount of fiber on the triaxial shear properties of the LSS with
different cement contents, is discussed.

Keyword: Liquefied Stabilized Soil, Slurry density, Cement content, Fiber Material, Triaxial Shear Property

1. INTRODUCTION 2. MATERIALS AND TESTING METHOD

Since the Liquefied Stabilized Soil (LSS) 1) is a cement (1) Test Materials
treated soil classified as a slurry premixed soil, it has
In this study, the New Snow Fine Clay (NSF-CLAY)
similar mechanical property as cement treated soil. With
was used as a homogenous base material, which is a
the increase of added amount of stabilizer, an increase in
commercially available cohesive soil with very clearly
strength is observed while the seismic resistance might
defined physical parameters shown in Table 1. Geoset 200
decrease due to its brittleness 2), 3). On the other hand, with
cement provided by Taiheiyo Cement Co. was used as
the aim of reducing the overburden pressure when the LSS
cement stabilizer, which is a cement-based solidifying
is used as a backfill soil for cut and cover tunnel etc., there
agent specially for soft clay and problematic soil.
is a concern about the use of the LSS prepared with a
Newspaper to be pulverized like cotton wool by a food
relatively lower slurry density instead of the normal slurry
processor was used as fiber materials (Fig. 1).
density with considering the strength reduction. In the past,
in order to improve the brittle property of the LSS, a series
of studies on the Liquefied Stabilized Soil mixed with
newspaper prepared like cotton wool as the fiber material
was carried out, and it was reported that the brittle
property after the peak is improved due to the reinforcing
effect of the fiber 4) ~6). However, there are few study cases
and still many unexplained parts about the LSS with Fig. 1 Pulverized newspaper
reduced slurry density7).
In this study, the influence of changing in slurry density
and cement content as well as added amount of fiber on
the triaxial shear property of the LSS is discussed.

Table 1 Physical parameter of NSF-CLAY

Particle density ρs (g/cm3) 2.762
Liquid limit WL (%) 60.15
Plastic limit WP (%) 35.69
Plasticity index IP 24.46
Fig. 2 Available range of slurry5)

1Student Member of IGS，Graduate Student, Muroran Institute of Technology (〒050-8585 27-1 Mizumoto-cho, Muroran)
2Member of IGS，Professor, Muroran Institute of Technology, Muroran, Japan (〒050-8585 27-1 Mizumoto-cho, Muroran)

- 123 -
(2) Mixing Method
In general, there are two LSS mixing methods used for
excavated soil, which are slurry type consisting of an
excavated soil, water and cement stabilizer, and
adjustment slurry type consisting of the contents of slurry
type LSS and fine-grained sand or clayey soil powder. In
this study, the slurry type was used due to easier
preparation. In this method, water was added moderately
to soil for adjusting a slurry density, then the cement
stabilizer was added and mixed. The general mixing test
of slurry type LSS were conducted with changing slurry
density and cement contents. The slurry density is defined
as the mass of slurry divided by the volume of slurry. The
slurry is prepared by mixing clay with water.
Thus, based on the result of flow test, breathing test and
unconfined compression test after 28-days curing, an
available range of slurry density as shown in Fig. 2 was
drawn with the flow value and unconfined compressive
strength. As shown in the figure, this range was carried out
with a cement content of 100 kg/m3 and the unconfined
compressive strength after 28-days curing. With the
available range of unconfined compressive strength of
200~500 kPa and flow value of 160~300 mm, the basic
slurry density in this study was decided to be 1.280 g/cm3,
and a changing rate of slurry density Df was defined as
(Actual slurry density)/ (Basic slurry density) ×100 %.
(3) Specimen Preparing Method
By applying the slurry type method, slurry with a target
density of 1.280 or 1.216 g/cm3 (in lower density case Df
= 95 %) was prepared by mixing water and NSF-Clay with
the amount calculated beforehand. After reaching the
target density by adjusting several times, cement was
added to the slurry with the amount of 80 and 100 kg/m3.
After adding the fiber material, the Liquefied Stabilized
Soil was fully mixed with a hand-type mixer. In order to
reduce the air bubbles that might be generated during
adding the fiber material and mixing, the LSS was
vacuumed in a sealed cell under a negative pressure of
about -90 kPa for 30 minutes. After removing the bubbles,
specimen was cured by the plastic mold with the size of
50 mm in diameter and 100 mm in height. Then after 28
or 56-days curing under constant temperature of around
20 ℃ and moderate moisture, triaxial compression test
was performed under undrained condition. All testing
cases are shown in Table 2.
Table 2 Details of testing cases
Cement content
80 100
(kg/m3)
1.216 1.216
Slurry density (g/cm3)
1.280 1.280
Fiber content (kg/m3) 0, 10
Curing time (days) 28, 56 28
Effective confining
98
pressure (kPa)
Axial strain rate
0.054
(%/min)
- 124 -
Fig. 3 Schematic figure of test
(4) Test Procedures
In this study, specimens with cement content of 80
kg/m3 were tested with both 28 and 56-days curing
condition while others with 100 kg/m3 were tested only
after 28-days curing.
As shown in the schematic figure in Fig. 3, the testing
system mainly consists of the axial loading system,
loadcell, pressure transducer (for cell and pore water
pressure), axial displacement transducer (external dial
gauge and local deformation transducer) and volume
change measurement (double tube burette). In order to
conduct small unloading/reloading loops, direct drive
motor (mega-torque motor) with high accuracy was
applied in the loading system and all progresses during
isotropic consolidation and triaxial shear were controlled
by the computer.
The measuring of axial strain in this study was carried
out by Local Deformation Transducer (LDT), which can
measure the axial displacement from small strain level
without the bedding error due to the compression of loose
layers at the top and bottom ends of specimen or filter
paper 8).
After setting the LDT with phosphor bronze hinge, the
specimens were saturated by the double vacuum method,
which vacuum pressure was applied and the de-aired
water flowed through specimen. After saturation, isotropic
consolidation was conducted for about 16 hours under the
effective confining pressure of 98 kPa. In order to study
the small strain deformation property, small
unloading/reloading loops during monotonic loading were
applied before the peak of the stress-strain curve. The
control condition of the unloading/reloading loops is as
follows. Until axial strain reach to 0.1 %, the
unloading/reloading loop was applied every axial strain
range of 0.01 %. After axial strain reach to 1 %, the loop
was applied every axial strain range of 0.1 %. The axial
strain rate was 0.054 %/min overall. This value was set by
reasons as follows. In order to record more data at small
strain level, the lowest speed available of the loading
system was applied. It is also identical to the author’s past
study.
Fig. 4 (a) Stress-strain relation (C = 80 kg/m3, 28-days)
Fig. 4 (b) Stress-strain relation (C = 80 kg/m3, 56-days)
Fig. 4 (c) Stress-strain relation (C = 100 kg/m3, 28-days)
3. RESULTS AND DISCUSSIONS
(1) Stress-Strain Relationship
Fig. 4 shows the relationship between deviator stress
and axial strain (LDT based) in the cases where the added
amount of fiber materials are 0 and 10 kg/m³ (Pc-0, Pc-
10) combined with slurry density of 1.216 g/cm³ (Df =
95 %) and 1.280 g/cm³ (Df = 100 %) of different cement
content at same curing time of 28-days respectively.
Fig. 4 (a) ~ (c) show that the maximum deviator stress
qmax is lower in the cases of Df = 95 % for all different
- 125 -
Fig. 5 Relationship between qmax and curing time
Fig. 6 Relationship between qmax and cement content
conditions of cement content and added amount of fiber
materials respectively.
Fig. 4 (a) and (b) indicate that the strength of the LSS
seems not to be affected by the curing time (28-days and
56-days) in this study. However, it is clearly shown by the
difference of Fig. 4 (a) and (c) that with the increase of
cement content, the strength of the specimens increases
quite noticeably. And the effect of slurry density seems to
become more obvious in the case of higher cement content.
Even in the case of low slurry density, it can be noticed
that the soil is improved by the fiber material.
Fig. 5 shows the relationship between maximum
deviator stress and curing time for cases with 80 kg/m3
cement content and Fig. 6 shows the relationship between
maximum deviator stress and cement content of all the
case with 28-days curing. With the increasing of curing
time, the strength of the LSS increases slightly. From Fig.
6, it is found that the qmax of 100 kg/m3 in case of Df =
95 % is larger than that of 80 kg/m3 in case of Df = 100 %.
Therefore, it is considered that the increase of cement
content is effective when the slurry density is decreased.
2) Deformation Property
a) Definition of various Young’s moduli
Fig. 7 shows the definitions of various Young’s moduli.
 Fig. 7 Definition of various Young’s moduli Fig. 10 (a) Etan/E0~q/qmax relation (C = 80 kg/m3, 28-days)

Fig. 8 Relationship between E0 and curing time Fig. 10 (b) Etan/E0~q/qmax relation (C = 80 kg/m3, 56-days)

Fig. 9 Relationship between E0 and cement content

Fig. 10 (c) Etan/E0~q/qmax relation (C = 100 kg/m3, 28-days)
The initial Young’s modulus E0 is defined as initial
stiffness at 𝜀a = 0.005 % or less. The tangent Young’s
deviator stress, the initial Young’s modulus is also
modulus Etan is defined as a tangential gradient in q𝜀a
affected by the changing of slurry density. However, the
curve, it indicates the non-linearity of deformation
increase in curing time and cement content did not has
property in q~𝜀a relation. The equivalent Young’s
such significant influence on the initial Young’s modulus
modulus Eeq is obtained from small unloading/reloading
as to be indicated for maximum deviator stress. In the past
loop during monotonic loading and which can indicate the
study, it is reported that the cement content can affect the
degree of damage under the shearing 9), 10).
initial Young’s modulus apparently at lower degree while
b) The initial Young’s modulus the influence decrease gradually along with the increase
Fig. 8 and Fig. 9 show the relationship between the of cement content 5).
initial Young’s modulus and the curing time as well as
cement content, respectively. Similar as the maximum

- 126 -
c) The tangent Young’s modulus
Fig. 10 (a)~(c) show the relationship between the
tangent Young’s modulus normalized by the initial
Young’s modulus (Etan/E0) and the deviator stress
normalized by the maximum deviator stress (q/qmax) in
different cases. These relations indicate the non-linearity
of pre-failure deformation property. The degree of non-
linearity can be compared normalizing the tangent
Young’s modulus and the deviator stress. It can be noticed
that even with the increase of curing time, the decreasing
tendency of Etan/E0 is always large in the cases of lower
slurry density as well as an increasing tendency of
nonlinearity. Meanwhile, by comparing Fig. 10 (a) and (c),
it can be found that the decreasing rate of Etan/E0 at the
early loading stage becomes lower with the increase of Fig. 11 (a) Eeq/E0~q/qmax relation (C = 80 kg/m3, 28-days)
cement content.

d) The equivalent Young’s modulus

The equivalent young’s modulus (obtained with small
loops of unloading/reloading during monotonic loading)
normalized by the initial young’s modulus (Eeq/E0) are
shown in Fig. 11 (a)~(c) with the horizontal axis of q/qmax.
With all the cases of different curing time and added
amount of fibers, Eeq/E0 seems to decrease earlier when
the slurry densities are relatively lower. Therefore, the
degree of damage with shear seems to be larger in the case
of lower slurry density, and that tendency becomes large
to be reduced with the addition of fiber materials.
As for the result obtained with higher cement content,
the change of cement content from 80 to 100 kg/m3 do not
have noticeable influence compared with that of curing Fig. 11 (b) Eeq/E0~q/qmax relation (C = 80 kg/m3, 56-days)
time.

4. CONCLUSIONS

The following conclusions can be obtained based on

test results.
1) When the slurry density is slightly decreased from
the normal slurry density obtained from the standard mix
proportion design figure, it is considered that the qmax
decreased remarkably. In addition, it is found that the local
damage caused by shearing even in the case of lower
slurry density is reduced by the fiber material.
2) The test result on specimen with higher cement
content indicated that cement content can affect the qmax Fig. 11 (c) Eeq/E0~q/qmax relation (C = 100 kg/m3, 28-days)
as same as the slurry density while the fiber material
seemed to behave well in these cases to improve the degree of damage tends to be reduced by the addition of
brittleness. fiber material.
3) The Etan/E0 ~ q/qmax relation showed that the 5) The effect of cement content on the stiffness is
decreasing rate of Etan/E0 in the early loading stage was relatively small in comparison with the effect of slurry
influenced by the cement content. In addition, it seems density, whereas the effect of cement content on the
that the addition of fiber materials can also decrease that strength is larger than the slurry density. Therefore, it is
rate slightly in all the cases. considered that even if a cement content increases, the
4) The influence of slurry density on the degree of decrease of slurry density causes the reduction of quality
damage caused by shear is quite large and it seems that the for the Liquefied Stabilized Soil.

- 127 -
ACKNOWLEDGMENTS
The authors express deep gratitude to Mr. Weichen Liu
(Graduate student, the University of Tokyo) for the help
in experimental works. The authors are also grateful for
the funding awarded by the Geo-Science Center
Foundation in 2017.
REFERENCES
1) Kuno, G., eds: Liquefied stabilized soil method-
Recycling technology of construction-generated soil
and mud, Gihodo publication, 1997 (in Japanese).
2) Japanese Geotechnical Society: Committee Report
Chapter 2, 2.1, 2.2 on test methods and physical
properties of cement-modified soil, Proc. of
symposium, pp.2-22, on survey, design, construction
and properties evaluation methods of solidifying
stabilized soil using cement and cement-treated soil,
2005 (in Japanese).
3) Kohata, Y.: Mechanical property of liquefied
stabilized soil and future issues, Doboku Gakkai
Ronbunshuu, F, Vol.62, No.4, pp.618-627, 2006 (in
Japanese).
4) Kohata, Y., Ichikawa, M., Nguyen, C. Giang., and
Kato, Y.: Study of damage characteristics of liquefied
stabilized soil mixed with fibered material due to
triaxial shearing, Geosynthetics Engineering Journal,
Vol.22, pp.55-62, 2007 (in Japanese).
5) Kohata, Y., Ito, K., and Koyama, Y.: Effect of cement
content on shear properties of liquefied stabilized soil
繊維材で補強された流動化処理土の強度・変形特性に及ぼす泥水密度
とセメント添加量の影響
YUJIE CUI・木幡行宏
本研究では，室内で養生された繊維材混合流動化処理土の強度・変形特性に及ぼす泥水密度とセメント添加量
の影響について検討している．実験条件として，基準泥水密度1.280 g/cm3及び1.280 g/cm3から5 %低減した泥水
密度で, セメント添加量を80, 100 kg/m3で, 繊維材を0, 10 kg/m3で混合した繊維材混合流動化処理土を作製した．
室内で28日養生された供試体に対して，それぞれ圧密非排水三軸圧縮試験（CUB試験）を実施した。試験結果よ
り，低泥水密度で作製された流動化処理土及び繊維材混合流動化処理土の品質が低下することが示された．一方，
土被り圧軽減を目的として基準泥水密度より小さい泥水密度で流動化処理土を作製する場合には，セメント添加
量を増加させることが有効であることが明らかにされた．
キーワード： 流動化処理土, 泥水密度, セメント添加量, 繊維材, 三軸せん断特性
- 128 -
reinforced by fiber material, Japanese Geotechnical
Society Hokkaido Branch Technical Report Papers,
Vol.26, pp.95-100, 2011 (in Japanese).
6) Duong, Q. Hung., Kohata, Y., Omura, S., and Ozaki,
K.: Strength and deformation characteristics of
liquefied stabilized soil reinforced by fiber material
prepared at laboratory and field, Geosynthetics
Engineering Journal, Vol.29, pp.33-40, 2014.
7) Do, T. Anh, Kohata, Y. and Sasaki, M.: Effect of slurry
density on triaxial compressive properties for
liquefied stabilized soil reinforced with fiber material,
Geosynthetics Engineering Journal, Vol.33, pp.15-22,
2018.
8) Goto, S., Tatsuoka, F., Shibuya, S., Kim, Y-S., and
Sato, T.: A simple gauge for local small strain
measurements in the laboratory, Soils and
Foundations, Vol.31, No.1, pp.169-180, 1991.
9) Kohata, Y., Tatsuoka, F., Wang, L., Jiang, G-L., Hoque,
E., and Kodaka, T.: Modelling the non-linear
deformation properties of stiff geomaterials,
Géotechnique, Vol.47, No.3, pp.563-580, 1997.
10) Kohata, Y., Jiang, L. G., Murata, O., and Tatsuoka, F.:
Elastic-properties-based modeling of non-linear
deformation characteristics of gravels, Proc. of the
second Inter. Symposium on Pre-Failure Deformation
Characteristics of Geomaterials-IS Torino 99, pp.533-
539, 1999.